Recent research in stem cell biology has been driven in large part by the immense potential of stem cells in the treatment of various human diseases. Given the absence of insulin production in Type 1 diabetes, the regeneration of pancreatic beta islet cells from stem cells could have direct clinical application for diabetic patients. However, neither the isolation nor production of pancreatic stem cells is currently a viable option for pancreatic regeneration. In order to overcome these obstacles, we propose to identify small molecules capable of inducing embryonic stem (ES) cells to differentiate into pancreatic stem cells. This will involve the screening of a 15,000 compound library for compounds that induce the expression of the homeobox protein PDX-1, an early marker of pancreatic stem cells. Subsequent assays will analyze for additional pancreatic markers, (i.e., neurogenin3 and insulin). Compounds meeting these criteria will be assayed for the ability to induce pancreatic tissue in eve in the chick embryonic assay. Small molecule-generated ES-derived "islet-like" cells will then implanted in NOD mice to determine if they are functional insulin-producing cells in vivo. We anticipate that the identification of a palette of pancreatic stem cell inducers will greatly stimulate basic research in this field (e.g., identification of additional pancreatic stem cell marker proteins via differential gene analysis).